Toward Something That Is Not Probability
Mathematical Futurology is Math but it is not Probability
For a long time I believed the problem with probability theory was foundational — that if you simply rebuilt it on stronger axioms, something truer would emerge. That intuition produced WMT, a probability framework built on Dependent Choice rather than full AC, defining probability values through sequential approximation rather than classical measure. It was rigorous. It was original. And ultimately I discarded it.
Not because it was wrong. Because it was too conservative. WMT was still in conversation with Kolmogorov. It was probability with better bones — a renovation, not a departure. The sample space survived. The scalar value survived. The basic ontological commitment — that uncertainty is something you quantify — survived untouched.
Mathematical Futurology begins by abandoning that commitment entirely.
The primitive object is a future — undefined, not a set, not an event. The valuation on futures is ternary: resolved, unresolved, and a third value that is neither. Not unknown. Not transitional. Constitutively unvalued — a property of the future itself, not of any observer’s knowledge. And the single foundational axiom is this: no future precedes itself.
A future cannot be a condition of its own existence. This blocks circular constitution at the foundation — the same pathology that destroys naive set theory, handled here before the framework is built rather than after.
No sample space. No measure. No probability.
What grows from these primitives will handle uncertainty — but the machinery will be foreign enough that calling it probability would be a category error. That is the goal. Not a better probability theory. Something that probability is a degenerate, flattened special case of.
We are at the beginning.